Abstract

The main theme of this dissertation is to explore how to establish a better relationship between sociological and philosophical investigations of science; how should epistemological considerations be used in sociological studies of scientific knowledge, and how should sociological findings be used in epistemological studies? To answer these questions, I review both sociological and social epistemological literatures on scientific knowledge, with more emphasis on the latter. On the sociological side, I point out that a large part of SSK literature is, contrary to what some philosophers believe, totally compatible with the idea of scientific rationality. Given that, I propose the 'even stronger' program of SSK that can better utilize the philosophical literature on scientific rationality. ;On the philosophical side, I review works of four philosophers in detail. I classify them into two general categories, the descriptive approach and the normative approach . After critical analyses of each of them, I apply their insights to a concrete case study, namely the issues related with diversification of Post WW II American sociology. These analyses and applications aim to show that all of them have some interesting things to say. One interesting observation in terms of the case study is that the diversity in theory and methodology in American sociology may be a good thing for various philosophical reasons, offered by these authors. ;Finally, the general characteristics of descriptive and normative approaches are compared. Some seeming incompatibilities between two approaches can be resolved by introducing some meta-epistemological notions. Especially the notion of WRE is important in identifying different roles played by two approaches. In any case, a good philosophical analysis cannot ignore sociological findings on actual scientific practices. To say the least, when there are two epistemological views, one of them does better than the other in accounting for actual scientific practices, this is a prima facie reason to prefer the one to the other